GMAT - Padlock on MBA Admissions Gates

Note: A printable PDF of this page is located here.

GMAT Basics:

The Graduate Management Admissions Test is the standardized exam used by graduate business schools for admissions decisions. It is designed and produced by the Educational Testing Service (ETS) and administered throughthe Graduate Management Admissions Council (GMAC). The three-hour, computer-adaptive test is almost exclusively multiple-choice and

GMAC has been able to validate the GMAT for just one purpose: predicting first-year graduate school grades. But GMAC concedes that the GMAT can predict less than 17% of the variation in these grades on average. Independent researchers put the percentage much lower at between 41 points before they indicate a difference in the abilities measured by the GMAT which means that colleges cannot determine the better qualified of two students whose scores are 530 and 570.

Admissions Cutoff Scores:

Due to the GMAT's weak validity and lack of precision, the GMAC Code of Ethics states: "Avoid the use of cutoff scores. Cutoff scores should only be used when there is clear empirical evidence that a large proportion of the applicants scoring below the cutoff scores have substantial difficulty doing satisfactory graduate work. GMAC continues, "In addition, it is incumbent upon the school to demonstrate that the use of cutoff scores does not result in the systematic exclusion of members of either sex, any age or ethnic groups, or any other relevant groups in the face of other evidence that would indicate their competence or predict their success." In addition, the National Association of Graduate Admissions Professionals' handbook Professional Standards and Practices states, "It is advisable to consider implementation of additional practices that eliminate discrimination, provide equal access and treat tests as a single component of the admissions procedure since test scores are not the sole measure of potential success."

Despite this strong guidance, hundreds of schools use absolute cutoff scores for admission and financial aid. For example, Texas Christian University has an accelerated MBA program that requires a 620 on the GMAT. Chapman College of Business and Concordia University each demand a 500, while California State University at Stanlihaus mandates a 450. The University of North Florida is even more explicit: "a minimum 20 verbal and 22 quantitative GMAT sub-score is required for graduate admission."

GMAC recognizes these abuses are widespread, noting among "Business School Application Tips" in one publication, "Minimum Score Requirements: Some schools require minimum scores on the Graduate Management Admission GMAT. Other schools only have minimum requirements for certain sections of the GMAT. Know the minimum requirement at a particular school, if one exists, so you only apply to schools you are qualified to attend."

GMAC takes no action against schools that misuse its test results. In fact, schools with cutoff scores are listed in GMAC's college search and financial aid identification programs and given all of the benefits of GMAC membership.

GMAT "Merit" Scholarships:

Millions of dollars in "merit" scholarships are based on GMAT cutoff scores. For example, the University of Oregon's Lundquist College of Business Scholarship requires a 650 on the GMAT. Many aid programs consider only the GMAT "Total" score (Verbal + Quantitative), potentially decreasing the number of women by excluding Analytic Writing.

The average cost to attend business school is a little over $10,000 per year. Top 25 schools however, cost more than double that, potentially pricing out many of the 95% of Black and 80+% of Latino test-takers who report they need financial aid. In contrast, Whites are least likely to need aid but most likely to get it from GMAT score-based "merit" scholarships.

Business School Rankings and the GMAT:

Rankings publicized by magazines such as U. S. News & World Report and Business Week contribute to the illusion that GMAT scores are a valid measure of the worth of graduate management programs and their students. The average GMAT scores of enrolled students count for almost one-sixth of a school's total weight in both publications. Nearly a quarter of all MBA applicants say these rankings are "extremely important" for deciding where to apply. Many business schools point to rising scores as evidence of higher standards, neglecting to put them in the context of the rapid climb of average GMAT scores -- 57 points in the past 18 years.

AVERAGE GMAT TOTAL SCORES 2000 - 2001*

*most recent year data is available

 

All Test-takers   527
Women   503
Men   541
Whites   538
Blacks   427
Latinos (several nationality groups)   451-474

 

Test Coaching Distorts Scores:

GMAT preparation courses from firms such as Princeton Review and Kaplan promise to raise scores by an average of 92 points for those who can afford $800 to $1000 or more. Thousands of college graduates eagerly shell out millions of dollars every year to these companies. Business school admissions offices have no way to tell which applicants have been coached and which have not been.

Women and the MBA:

Women graduate from college with higher undergraduate Grade Point Averages (GPAs) than men but score 38 points lower on the GMAT- - a gap that has grown by 28 points since 1982. Women who graduate from Business School have, on average, GPA's equivalent to those of men. Independent research indicates that the GMAT over-predicts the performance of men and under-predicts that of women.

Test scores play a role in school choice. Men are much more likely than women to send their GMAT reports to "Top 20" schools and define such a school as their top preference. Women submit fewer applications and apply to less selective schools than their male counterparts.

The number of women in business schools peaked at 30% and remained flat for several years. GMAC reports that while 87% of MBA programs received more total applications for 2001-2002, less than half of these schools saw an increase in applications from women. Almost one-third of business schools at private universities have seen a drop in female applicants.

African Americans and Latinos:

In 2000-2001 African Americans scored, on average, 111 points lower than Whites. Latino populations scored between 64 and 87 points lower than Whites. The huge score gap between Blacks and Whites has narrowed only slightly in two decades, while the difference between Latinos and Whites has widened.

In 1994-1995 only 143 Blacks scored higher than 650 (over 100 points lower than the averages for top-tier schools). High-scoring Whites outnumbering high- scoring Blacks 75 to 1. Without affirmative action, African Americans and Hispanics would be much less likely to get into a top-25 graduate business school.

Members of these groups enter the MBA pipeline with far more concerns regarding their abilities to meet academic and curriculum demands, fears reinforced by much lower GMAT scores. African Americans and Hispanics are greatly disadvantaged by a test with little predictive ability and multiple opportunities for abuse.

Other Groups:

Applicants who speak English as a second language (ESL) are disadvantaged by the primarily multiple-choice nature of the GMAT and the fast pace at which items must be answered. ETS studies show that when these characteristics are not part of the test (as in the Analytic Writing section) the score gap between whites and ESL applicants is greatly reduced.

GMAC admits that average scores decline for test-takers over the age of 31. There is also an inverse relationship between GMAT scores and business experience. Studies conducted for GMAC show GMAT scores under-predict the performance of older students (particularly women) but demonstrate that business experience directly contributes to success in MBA programs.

While little research has been done on the validity of GMAT scores for students with disabilities, the test is likely to have the same discriminatory impact and low predictive validity as similar standardized admission tests produced by ETS such as the SAT.

Alternatives:

GMAC and ETS have long been the national gatekeepers for business school admissions. Soon there might be some competition. Researchers at the University of Michigan's School of Business have developed alternative measures of managerial potential that they claim better predict success in both academic and practical endeavors. This new assessment approach, dubbed the "Rainbow Project," also claims to produce far less racial and gender bias than the GMAT.

In 1985 Harvard Business School (HBS) decided to eliminate the GMAT from their admissions process. John Lynch, the Admissions Director at the time, gave several compelling reasons. In a blind test, Harvard found that admissions decisions made with and without the GMAT were essentially the same. Success at Harvard depended on intangibles such as motivation, interpersonal skills, perseverance and hard work – all factors not measured by GMAT. Looking at undergraduate grade-point average (UGPA), ethics, leadership, community activities, prior work experience and the interview made GMAT scores "superfluous". Harvard was also concerned about the perceived emphasis applicants place on the GMAT and that strong applicants with scores below the 99th percentile were intimidated from applying. Lynch also pointed out that an "artificial barrier to the admission of qualified but poorer students is unacceptable."

Eleven years later Harvard reinstated the GMAT, saying the decision "reflects the fact that there have been significant improvements to the structure of the test". The only changes made since 1954 have been the addition of the AWA, which is used only sporadically and the move to the Computer Adaptive Test, which came with multiple problems. Harvard also pledged to work closely with GMAC to cooperate "on ways to further change the test" to focus on intangible qualities. In six years, no further mention has been made of changes to the test.

Guide to Making Changes

Schools that have de-emphasized standardized tests in admissions have done so for a wide range of reasons and have chosen several different approaches. The following steps serve as a guide:

  1. Conduct a standardized test "audit" to understand how your school uses the GMAT and clarify current test uses for admissions and scholarship awards. Evaluate these goals in relationship to the overall mission of the school. Consider broader admissions issues, including diversity, outreach, access and cost.
  2. Conduct an independent validity study and compare your experience with the GMAT with the validation and correlation studies conducted by GMAC. GMAT scores provide useful information only to the extent that they lead admission officers to make different- and better- decisions than they would make without the test.
  3. Assess the impact of a variety of admissions requirements on different applicant groups such as women, Latinos, African Americans, older applicants and students with disabilities.
  4. Evaluate the predictive value of alternate criteria that influence college persistence and success such as UGPA, work experience and motivation.
  5. Evaluate the impact of GMAT requirements and/or minimums on self-selection, self-efficacy and school choice- particularly on underrepresented groups.
  6. Develop admissions alternatives, including: make submission of GMAT scores optional; require the GMAT only for those applicants with UGPAs below a certain point; and drop the GMAT completely.

National Center for Fair & Open Testing (FairTest)
15 Court Square, Suite 820, Boston, MA 02108
(857) 350-8207
www.fairtest.org
university@fairtest.org

FairTest GMAT Bibliography

2002 Graduate Management Admission Council (GMAC) Applications Trends Survey. www.gmac.com, March 2003.

2002 Global MBA Survey. GMAC Web Site, www.gmac.com, March 2003.

2002 Corporate Recruiter Survey Results. GMAC web site, www.gmac.com, March 2003.

An Admissions Office Profile of Candidates Taking the GMAT, 1982-1983 through 1986-1987. GMAC, June 1988.

An Admissions Office Profile of Candidates Taking the GMAT, 1996-1997 through 2000-2001. GMAC, 2002.

Baydar, Nazli, Grady, William R., Dugan, Mary K. and Johnson, Terry R. The GMAT Registrant Survey Wave II Report: And Early Look at GMAT Registrants' Progress Towards an MBA Degree. GMAC, 1995.

Benson, Gary. GMAT–Fact or Fiction: A Look at the Validity of the Exam. Paper presented at the Annual Meeting of the Rocky Mountain Educational Research Association, Tucson, AZ, November 2-5, 1983.

Bridgeman, Brent and McHale, Frederick. Gender and Ethnic Group Differences on the GMAT Analytical Writing Assessment, Educational Testing Service, February 1996.

Bruce, Grady D. Customer Satisfaction with GMAC: Results of a Survey of Admissions Directors at GMAC Schools. GMAC, Nov 1997.

Bruce, Grady, D. Customer Satisfaction with GMAC: Results of a Survey of Deans in GMAC Schools. GMAC, May 1998.

Dugan, Mary K., Klepinger, Daniel H., and McLaughlin, Steven D. Entering the MBA Pipeline: Examining the Factors Affecting the Intention to Seek a Graduate Business Degree. GMAC Selections Magazine, undated.

Edgington, Rachel and Olkin, Jacqueline. Who Gets in and Why. GMAC Selections, Spring 2002.

Bruce, Grady D. Customer Satisfaction with GMAC: Results of a Survey of Admissions Directors at GMAC Schools. GMAC, Nov 1997.

Grady, William R., Johnson, Terry R., Dugan, Mary Kay and Baydar, Nazli. Program and Gender Differences in Progress through the MBA Pipeline: Wave III of the GMAT Registrants Survey. GMAC Selections, undated.

Gruber, Betsy, Littman, Margaret and Merritt, Jennifer. Business Week's Guide to the Best Business Schools. McGraw Hill, 7th Edition 2001.

Guide to the Use of GMAT Scores. GMAC web site, www.gmac.com, October 2002.

Hancock, Terence. The Gender Difference: Validity of Standardized Admission Tests in Predicting MBA Performance. Journal of Education for Business, v. 75. no. 2, 1999.

Hecht, Lawrence W., Manning, Winton H., Swinton, Spencer S., with Braun, Henry I. Assessing Older Applicants for Admission to Graduate Study in Management: The Role of GMAT Scores and Undergraduate Grades. GMAC, November 1989.

McLaughlin, Steven D. and Johnson, Terry R. The GMAT Registrants Survey Wave I Report: Characteristics of GMAT Registrants and Their Likelihood of Taking the Test. GMAC, 1993.

McLaughlin, Steven D. and Johnson, Terry R. Entering the MBA Pipeline: A Preliminary Look at Findings from the GMAT Registrants Survey. GMAC Selections Magazine, Spring, 2002.

Montgomery, Mark. A Nested Logit Model of the Choice of a Graduate Business School. Grinnell, IA: Grinnell College, 2001.

Montgomery, Mark, Johnson, Terry R. and Dugan, Mary K. How Much Do Minorities Value Affirmative Action in Admission Decisions? Evidence from Graduate Business School. Grinnell, IA: Grinnell College, 2002.

Palmer, John C. and Wright, Robert E. "The Influences of Student Age and Gender on the Predictive Validity of GMAT scores and Undergraduate GPAs," Educational Research Quarterly, Vol. 23, No. 2, 1999, p.27-33.

Validity Study Service Quick Facts/FAQ's. GMAC web site, www.gmac.com, February 2003-April 2003.>

Wright, Robert E and Palmer, John C. GMAT Scores and Undergraduate GPAs as Predictors of Performance in Graduate Business Programs. Journal of Education for Business, v69 n6 p344-48 Jul-Aug 1994.

Wright, Robert E., and Palmer, John C. "Examining Performance Predictors for Differentially Successful MBA Students," College Student Journal, Vol. 31, #2, 1997, p. 276-281.

Wright, Robert E., and Palmer, John C. Predicting Performance of Above and Below Average Performers in Graduate Business Schools: A Split Sample Regression Analysis. Educational Research Quarterly, Vol. 22, #4, 1999, p. 35-44.

Xiao, Beiling. Factors Influencing Master's Degree Attainment in Business, Engineering, Health and Human Sciences, and Visual and Performing Arts. AIR 1998 Annual Forum Paper. ED424807

Zwick, Rebecca. The Validity of the GMAT for the Prediction of Grades in Doctoral Study in Business and Management: An Empirical Bayes Approach. Journal of Educational Statistics, v18 n1 p91-107 Spring 1993.

A printable PDF version of this factsheet is available here.